toimmune hemolytic anemia. The ELISA was more sensitive and correlated with severity of hemolysis better than the direct antiglobulin test (DAT). It was helpful ...
The Enzyme-linked Immunosorbent Assay: Accurate Detection of Red Blood Cell Antibodies in Autoimmune Hemolytic Anemia DAVID BODENSTEINER, M.D., PEGGY BROWN, PH.D., BARRY SKIKNE, M.D., AND FRED PLAPP, M.D., PH.D.
The enzyme-linked immunosorbent assay (ELISA) was employed in the study of red blood cells from patients with autoimmune hemolytic anemia. The ELISA was more sensitive and correlated with severity of hemolysis better than the direct antiglobulin test (DAT). It was helpful in diagnosing and following the clinical course in these patients. This was particularly true in the DAT-negative group, since the ELISA can detect smaller increases in red blood cell IgG than are required for a positive DAT. (Key words: Enzyme-linked immunosorbent assay; Red blood cell antibody; Autoimmune hemolytic anemia; Direct antiglobulin tests; DAT-negative hemolytic anemia; Enzyme-linked antiglobulin test) Am J Clin Pathol 1983; 79: 182-185 THE A U T O I M M U N E HEMOLYTIC ANEMIAS (AIHA) are a complex group of disorders resulting in premature destruction of red blood cells.8 This destruction is caused by immunoglobulins or other proteins directed against antigens on the red blood cell surface.6 IgG alone, or in combination with complement, is responsible for the majority of these cases. The IgG can be detected by the direct antiglobulin test (DAT), which is usually positive when 300-500 IgG molecules per red blood cell (IgG/RBC) are present. However, some cases with clinical and laboratory evidence of AIHA have a persistently negative DAT.2-7 Using a more sensitive technic, complement fixing antibody consumption, Gilliland and associates2 have shown that the red blood cells in these patients do have increased numbers of IgG molecules on their surfaces. Recently, even more sensitive technics were reported. Szymanski and co-workers7 described an automated antiglobulin test using an autoanalyzer, which can detect 100-150 IgG molecules per RBC. Galili and colleagues1 increased the sensitivity of the DAT as much as 50-fold using erythrocyte antibody (EA) rosette formation. These latter technics have not been widely used, because they are difficult and require special equipment. Received May 10, 1982; received revised manuscript and accepted for publication June 18, 1982. Presented at the Kansas American College of Physicians Meeting, Topeka, Kansas, February 1981, and at the American Association of Blood Banks, Chicago, Illinois, October 1981. Address reprint requests to Dr. Bodensteiner: University of Kansas, Department of Internal Medicine, College of Health Sciences, 39th Street and Rainbow Boulevard, Kansas City, Kansas 66103.
Divisions of Hematology and Pathology, University of Kansas Medical Center, Kansas City, Kansas
Alternatively, we have employed an enzyme-linked immunosorbent assay (ELISA) to determine the IgG content of red blood cells from patients with AIHA who had both positive and negative DATs. This test is easy to perform. The results are reproducible, and the sensitivity is greater than with the DAT. We have found a strong correlation between the clinical course, the hematologic profile, and the number of IgG molecules per red blood cell as determined by the ELISA. Materials and Methods Blood samples from healthy volunteers and patients with hemolytic anemia were collected in sodium ethylenediaminetetraacetate (EDTA). The washed red blood cell suspensions were counted by an ELT-8 laser hematology counter (Ortho Instruments, Westwood, MA). All cells were studied within one week of collection. Rh 0 (D) Immune Globulin (RhoGAM Lot RHV403) was obtained from Ortho Diagnostics (Raritan, NJ). The alkaline phosphatase anti-human IgG (Lot 61-270 S759) was from Miles Yeda Ltd (Israel), and the para-nitrophenyl phosphate (Sigma 104 phosphatase substrate tablets) was from Sigma Chemical Co. (St. Louis, MO). All absorbances were measured on a Gilford model 250 spectrophotometer (Gilford Instrument Laboratories, Oberlin, OH). Polypropylene tubes (12 X 75 mm) and a 0.2% Pentex bovine albumin-phosphate buffered saline (pH 8) solution (PBS-BSA) were used throughout. Outdated Rh-positive RBCs were obtained from the University of Kansas Medical Center Blood Bank. RBC phenotypes were determined according to American Association of Blood Banks guidelines. Control and patient RBCs were washed in phosphate-buffered saline and adjusted to a cell count of 5 X 108 RBC/mL. RBC Enzyme-linked
Immunosorbent
Assay
The ELISA was performed according to Leikola and Perkins.3 The standard curve (Fig. 1) was prepared by
0002-9173/83/0200/0182 $01.00 © American Society of Clinical Pathologists
182
Vol. 79 • No. 2
RED BLOOD CELL ANTIBODY DETECTION
183
incubating0.5 mL of 1:1000 to 1:128,000 serial dilutions of Rh 0 (D) Immune Globulin with 0.2 mL of control RBCs for 1 hour at 37°C in polypropylene tubes. The anti-D was diluted in PBS-5% BSA. Triplicate tubes containing 0.2 mL of test RBCs were suspended in 0.5 mL of PBS-5% BSA. Following incubation, the control cells and test cells were washed four times in PBS-0.2% BSA by centrifugation at 3,000 g for 5 minutes. The supernatants were completely decanted, and 200 /xL of a 1:400 dilution of alkaline phosphatase conjugated goat antihuman IgG were added to each tube and shaken at 37 °C for 30 minutes. RBCs were then washed four times, and the supernatants were decanted. Each tube was incubated with 200 jtL of 1 mg/mL para-nitrophenyl phosphate dissolved in 60 mM N a H C 0 3 (pH 9.6) for 1 hour at room temperature. The tubes were centrifuged, and 100 fiL of each supernatant was added to 200 nL of 0.1 N NaOH. The absorbances were measured at 405 nm in a Gilford Model 250 spectrophotometer. Standard Curve Calculations The number of IgG molecules per RBC for the serial dilutions of the standard curve were calculated in the following manner. Since one vial of Rh 0 (D) Immune Globulin (Lot RHV403) contains 346 ng anti-D/mL, 0.5 mL of 1:1000 dilution will have 173 ng anti-D/mL. Tube 1 of the standard curve has 0.5 mL of 1:1000 antiD and 1 X 108 RBCs. 173 X 10~9g anti-D 1.5 X 105g/mole = 115 X 10"14 moles anti-D in Tube 1. (115 X 10~'4 moles)(6 X 1023 molecules/mole) = 692 X 109 molecules anti-D. 692 X 109 molecules 1 X 108 RBCs = 6,920 molecules anti-D/RBC in Tube 1. Values continue as seen in Figure 1. A new standard curve is prepared each time the ELISA is done. Patients Six patients with AIHA, four DAT positive and two DAT negative, were studied. The controls consisted of normal volunteers and patients with cold agglutinin disease; microangiopathic hemolytic anemia; hereditary spherocytosis; pyruvate kinase deficiency; G6PD deficiency; Waldenstrom's macroglobulinemia with complement-mediated hemolysis; treated B| 2 , folate, or iron deficiency; and IgG myeloma. The patients with myeloma had more than 3 g of IgG per deciliter.
1-1000 1-2000
1-8000
1-32000
1-128000
Dilutions FIG. 1. Dilution series of anti-D-coated red blood cells plotted against optical density of final color change with the corresponding IgG/RBC values. Case I. A 28-year-old woman developed a DAT-positive hemolytic anemia in 1978 and was first seen by us in September 1979. Her hemoglobin was 10 g/dL, reticulocyte count was 28%, DAT was 3+ for IgG and negative for complement, and the IgG/RBC was 1,300. She did not improve while taking 100 mg prednisone daily. Following splenectomy, in the absence of steroids, the hemoglobin rose to 14.7 g/dL, the reticulocyte count dropped to 4.4%, the IgG/RBC dropped to 950, and the DAT remained positive. She is doing well but continues to hemolyze. Case 2. A 54-year-old man was referred for evaluation of a hemolytic anemia in September 1979. His hemoglobin was 9.7 g/dL and the reticulocyte count was 18%. The DAT was 1 + for IgG and negative for complement, and the IgG/RBC was 1,300. Treatment was begun with 60 mg prednisone daily. One month later, the hemoglobin had risen to 14.2 g/dL and the reticulocyte count was 1.6%. By January 1980, the hemoglobin was stable at 14.2 g/dL, the DAT was negative, and the IgG/RBC was 54. Case 3. A 55-year-old woman presented with a DAT-positive hemolytic anemia in February 1980. The hemoglobin was 5.8 g/dL, reticulocyte count was 18%, DAT was 3+ for IgG, and 1+ for complement. The IgG/RBC was 26,000. She has been treated with a gradually decreasing dose of prednisone and now, with 10 mg prednisone daily, has a hemoglobin of 12.2 g/dL and a reticulocyte count of 4.8%. The DAT remains positive and the IgG/RBC has dropped to 4,000. Case 4. A 74-year-old man with angioimmunoblastic lymphadenopathy (AILD), presented in January 1980 with a hemoglobin of 5.9 g/dL, reticulocyte count of 22%, DAT 3+ for IgG and trace for complement, and an IgG/RBC of 1,080. With prednisone, his hemoglobin rose to 10 g/dL and the IgG/RBC was 900 after 2 weeks. Two months later while taking no steroids, his hemoglobin was 12.6 g/dL, reticulocyte count 1.6%, DAT negative, and the IgG/RBC less than 54. In July 1980, the IgG/RBC had risen to 475, but the hemoglobin was
BODENSTEINER ET AL.
184
Table 1. IgG Molecules per Red Blood Cell in Conditions other Than IgG-mediated AIHA
A.J.C.P. • February 1983
Results
Disorder
Number
IgG/RBC
Normal volunteers Cold agglutinin disease Microangiopathic hemolytic anemia Hereditary spherocytosis Pyruvate kinase deficiency G6PD deficiency Waldenstrom's macroglobulinemia IgG myeloma IgG myeloma
25 1 3 1 2 1 2 4 1
